First test on DIY ebike running slow and hot

W

wobble

10 mW
Joined
Jul 23, 2020
Messages
25
Hello, this is my first ebike build so please excuse any rookie errors.

I just took it out for its first ride and its incredibly slow. The controller got quite warm too.
My throttle displays 4 bars for 'battery life' (but i assume this just read the voltage) and it dropped down to 2 bars whilst testing it.

I'd like to make it go faster, but i am not sure how to. I think i should be getting more speed out of it than i am.

My components are:
-36V controller rated for 800W (current limit 28A)
-5V throttle
-self built 10s4p Li-ion battery pack with 36V 40A BMS
-500W 36V brushed motor, however it is rated for 26.7A so do this mean 36*26.7=961.2W ?

I would greatly appreciate any advice/help.
Cheers

this is the controller: https://www.banggood.com/36V-800W-Brushed-Speed-Controller-Throttle-Twist-Grip-For-Scooter-Motor-E-Bike-p-1105937.html?utm_design=18&utm_email=1593160683_2332&utm_source=emarsys&utm_medium=Neworder171109&utm_campaign=trigger-order&utm_content=linux&sc_src=email_2675773&sc_eh=b6e06a1e9cada5e51&sc_llid=23331960&sc_lid=105229698&sc_uid=7qRUHco3E8&cur_warehouse=CN
 
whats your no load amp draw, sounds like wrong combination
 
I couldn't take an accurate reading due to my mutlimeter maxing out at 19 A, but it goes above this with no load.
Which components are wrong?
 
whoa! it should be waaaaaay lower. let someone else chime in, or search for phase/hall combo on this site
 
You mention your battery is configured 10S/4P - what is the total Amp hour capacity?
 
EDIT: My apologies; the below is only for brushless motors and does not apply here. :oops:


If the controller has self-learn wires, get the wheel off ground, then cnnect them. IF the wheel spins forwards, disconnect them, power cycle the bike, and go ride. If it spins backwrds, disconnect and reconnect, then it'll spin forwards, and you're set.

Otherwise:
https://endless-sphere.com/forums/viewtopic.php?f=2&t=70947&hilit=phase%2A+hall%2A#p1070826
John in CR said:
To protect the system I use a small jumper wire to complete my battery connection, and use only small short throttle pulses. The jumper can't pass high enough current to hurt anything, so it acts like a fuse, and the small pulses protect the jumper from overheating.

Understand that every phase combo has 1 valid hall combo, and every hall combo has 1 valid phase combo. 3 of the valids are forward and 3 are reverse.

Then decide which is easier to swap around, halls or phases, and keep the other set static. Try all 6 combos with just a small short pulse of throttle. That will give you one valid combo, but beware of false positive results which have more startup noise and generally spin the opposite direction of the 1 valid combo.

If the nice smooth running valid is forward, then you're done, though it's good to measure no load current at WOT to be sure with proper connections wired up, not the jumper.

If the valid one is reverse, then swap 2, any 2, only 2 of the wire you were keeping static and go back to the originals and find the valid smooth forward.

To be systematic when swapping, only change 2 at a time. It's all very simple. I've done it a number of times with wires all the same color, and I've also done it with motors having 6 phase wires and 10 hall wire. 3 phase is a piece of cake. Just relax and take your time. Alligator clip jumper wires make the process simple and quick.

Take care to avoid shorts, and if you have the chain on beware of reverses so the pedals don't whack you or get tangled in the wiring.
Click to expand...




Other examples are in some of these threads, and others not listed (can be found under other terms like wir* comb*).
https://endless-sphere.com/forums/search.php?keywords=phase*+hall*&terms=all&author=&sc=1&sf=titleonly&sr=topics&sk=t&sd=d&st=0&ch=300&t=0&submit=Search
 
My motor is brushed
https://www.amazon.co.uk/gp/product/B076H7QLR7/ref=ppx_yo_dt_b_asin_title_o06_s00?ie=UTF8&psc=1
 
amberwolf said:
If the controller has self-learn wires, get the wheel off ground, then cnnect them. IF the wheel spins forwards, disconnect them, power cycle the bike, and go ride. If it spins backwrds, disconnect and reconnect, then it'll spin forwards, and you're set.
Click to expand...

Does this apply for brushed motors?
 
E-HP said:
amberwolf said:
If the controller has self-learn wires, get the wheel off ground, then cnnect them. IF the wheel spins forwards, disconnect them, power cycle the bike, and go ride. If it spins backwrds, disconnect and reconnect, then it'll spin forwards, and you're set.
Click to expand...

Does this apply for brushed motors?
Click to expand...

Im not sure, his quote was referring to hall sensors which i know are for brushless motors
 
Brushed motor has no hall sensor or phase wires. Simply two wires, a pos and a neg. That much amperage suggests either shorted windings or a shorted controller. Brushed controllers are cheap, I would try another one.
 
My apologies, I couldn't open the link (page comes up blank white then redirects to the homepage),and missed the info in the OP about it being brushed. :oops:

So none of my advice in my first reply applies (but would if it were brushless).

Here's some brushed advice:


For brushed motors that have that high a no-load current, it means something in the system is applying a very high load to the motor, like really excessive chain tension, or something rubbing on a wheel or rim (frame, brakes, etc), or something is wrong with the motor itself (probably not the controller, or it likely would have failed completely under load, or be stuck on, so that the throttle doesn't control it, it just goes full on as soon as you power it up). It could also be that the gearing is completely wrong, far too "tall", but that typically doesn't make *that* much difference with no load, only when riding it.

So, first thing is disconnect the chain from the motor, and run it, measuring the current from the battery. If it drops dramatically, like down to an amp or so, then the motor and system itself is ok, and it's something mechanical.

If it doesn't drop, then most likely the motor has something wrong, which could be anything from a poorly installed brush/holder on the armature, which may be fixable, to a shorted winding, which is not. If the brushholder is rotatable, it's possible to advance or retard the timing of the motor from the normal neutral, and if it's too far off, it'll draw excessive current even in no-load. If it's riveted to a coverplate, but the plate can be installed in more than one rotation, and is put on the wrong way, it can cause that. Probably not to the degree you see, though.
 
amberwolf said:
My apologies, I couldn't open the link (page comes up blank white then redirects to the homepage),and missed the info in the OP about it being brushed. :oops:

So none of my advice in my first reply applies (but would if it were brushless).

Here's some brushed advice:


For brushed motors that have that high a no-load current, it means something in the system is applying a very high load to the motor, like really excessive chain tension, or something rubbing on a wheel or rim (frame, brakes, etc), or something is wrong with the motor itself (probably not the controller, or it likely would have failed completely under load, or be stuck on, so that the throttle doesn't control it, it just goes full on as soon as you power it up). It could also be that the gearing is completely wrong, far too "tall", but that typically doesn't make *that* much difference with no load, only when riding it.

So, first thing is disconnect the chain from the motor, and run it, measuring the current from the battery. If it drops dramatically, like down to an amp or so, then the motor and system itself is ok, and it's something mechanical.

If it doesn't drop, then most likely the motor has something wrong, which could be anything from a poorly installed brush/holder on the armature, which may be fixable, to a shorted winding, which is not. If the brushholder is rotatable, it's possible to advance or retard the timing of the motor from the normal neutral, and if it's too far off, it'll draw excessive current even in no-load. If it's riveted to a coverplate, but the plate can be installed in more than one rotation, and is put on the wrong way, it can cause that. Probably not to the degree you see, though.
Click to expand...

Thank you thats really helpful information!
I just got out the multimeter and the motor by itself when unloaded is drawing over 19 amps. I then tested the current supplied by the controller, giving a very slight turn of the throttle and that gave out more than 19 amps.
Is it possible that the motor is not designed for this kind of use? It is sold as a go-kart motor but that could be false
 
Also the voltage supplied is a contant value of 35v at all throttle degrees of rotation
 
You could try to connect the motor directly to the battery and measure the no load amp to rule out the controller being at fault. Just be careful, it will start pretty violently.

I have installed three of those motors (just the 800w versions) on different kids toys. It pushes 50-60kg without heating up at all on level ground as long as its geared correctly.
 
HrKlev said:
You could try to connect the motor directly to the battery and measure the no load amp to rule out the controller being at fault. Just be careful, it will start pretty violently.

I have installed three of those motors (just the 800w versions) on different kids toys. It pushes 50-60kg without heating up at all on level ground as long as its geared correctly.
Click to expand...

Ahh this is really good to know, so i guess the problem is with the controller.
What kind of gearing did you use?
Do i need more torque with this kind of motor?
 
wobble said:
Also the voltage supplied is a contant value of 35v at all throttle degrees of rotation
Click to expand...
35v from the throttle?

Or 35v from the controller to the motor?

If the former, something is seriously wrong with the wiring, and the controller could be fried. If you're using a pot throttle, it should vary about 0v to about 5v. If you're using a hall throttle, it should vary about 1v to about 4v.

If the latter, remember that the controller PWMs the battery voltage so every pulse is going to be about battery voltage, then down to zero, then back up to battery voltage, but they should be so fast your meter reads them as an average voltage, which should vary from zero to about battery voltage, over the full input range of the throttle.

If it does not vary with throttle input, but the throttle voltage does vary, there could be a miswiring so the throttle voltage isn't going to the throttle signal, but the wire from the controller for that signal is wired somewhere else providing it a nearly full-on voltage. If that's the case, it will also turn on and run the entire time it's powered on, and won't matter what the throttlle position is. The same is true if the controlelr FETs are shorted (common failure mode).

If the throttle voltage doesnt' vary at all, it's probably miswired or defective, and again it will also always run the motor the entire time it's powered on.

If there is some other thing that is happening, you'll need to describe it *exactly*, and provide the measurements made at the throttle signal wire in reference to throttle ground wire.
 
wobble said:
Ahh this is really good to know, so i guess the problem is with the controller.
Click to expand...
Does this mean that it operates normally when hooked directly to the battery?

If not, it's still the motor.

I don't see any way the controller in a brushed system can possibly cause the motor, when unloaded, no chain on it, to heat up or take lots of current. It requires a problem in the motor itself for that to happen.
 
amberwolf said:
wobble said:
Also the voltage supplied is a contant value of 35v at all throttle degrees of rotation
Click to expand...
35v from the throttle?

Or 35v from the controller to the motor?

If the former, something is seriously wrong with the wiring, and the controller could be fried. If you're using a pot throttle, it should vary about 0v to about 5v. If you're using a hall throttle, it should vary about 1v to about 4v.

If the latter, remember that the controller PWMs the battery voltage so every pulse is going to be about battery voltage, then down to zero, then back up to battery voltage, but they should be so fast your meter reads them as an average voltage, which should vary from zero to about battery voltage, over the full input range of the throttle.

If it does not vary with throttle input, but the throttle voltage does vary, there could be a miswiring so the throttle voltage isn't going to the throttle signal, but the wire from the controller for that signal is wired somewhere else providing it a nearly full-on voltage. If that's the case, it will also turn on and run the entire time it's powered on, and won't matter what the throttlle position is. The same is true if the controlelr FETs are shorted (common failure mode).

If the throttle voltage doesnt' vary at all, it's probably miswired or defective, and again it will also always run the motor the entire time it's powered on.

If there is some other thing that is happening, you'll need to describe it *exactly*, and provide the measurements made at the throttle signal wire in reference to throttle ground wire.
Click to expand...

My apologies 35v from the controller to motor, which remains constant
 
amberwolf said:
wobble said:
Ahh this is really good to know, so i guess the problem is with the controller.
Click to expand...
Does this mean that it operates normally when hooked directly to the battery?

If not, it's still the motor.

I don't see any way the controller in a brushed system can possibly cause the motor, when unloaded, no chain on it, to heat up or take lots of current. It requires a problem in the motor itself for that to happen.
Click to expand...

The motor has never heated up, its the controller which is heating up. The controller is supplying over 19A of current to the motor in an unloaded state.

The motor runs when connected to the battery, but im not able to test it loaded without it ripping the chain off of the sprocket instantly. Once i get the correct bolts i can secure the motor properly
 
But what is the actual throttle signal voltage, at the controller's signal input?

If that dosent' vary, then neither will the controller output.
 
amberwolf said:
But what is the actual throttle signal voltage, at the controller's signal input?

If that dosent' vary, then neither will the controller output.
Click to expand...

The throttle was about 4.4V and it is able to vary the speed of the motor. The seller of the controller had paired it with the throttle and sold the two together in a package deal.
 
wobble said:
So the throttle sohuld vary voltage and not current?
Click to expand...
The throttle itself always varies it's voltage.

What it causes the controller to vary at the motor, for a brushed system, is the width of the voltage pulse to the motor. The voltage pulse is always battery voltage. How long it applies that voltage determines the current draw of the motor. The current is also in pulses, becuase the voltage is in pulses, and also changes. So the controller varies the average current by varying the average voltage by varying the width of full-battery-voltage pulses. Because of the motor inductance and resistance, it may not look like "squarewave" pulses, but that's how brushed controllers work.

Unless you use an oscilloscope you can't see this occuring, and the only "valid" multimeter / ampmeter / voltmeter measurement you can make of what the motor is doing is at the battery input to the controller, not at the motor output from the controller.


wobble said:
amberwolf said:
But what is the actual throttle signal voltage, at the controller's signal input?

If that dosent' vary, then neither will the controller output.
Click to expand...

The throttle was about 4.4V and it is able to vary the speed of the motor. The seller of the controller had paired it with the throttle and sold the two together in a package deal.
Click to expand...
If it's always 4.4v, then it cannot vary the speed of the motor.

The throttle voltage must vary with throttle rotation.

If it is able to vary the speed of the motor, then the voltage at the motor, and the current draw from the battery, will also vary.

What is the actual throttle signal voltage, at the controller's signal input, when at zero rotation?

What is the actual throttle signal voltage, at the controller's signal input, when at maximum rotation?

What is the motor's current draw when directly connected to the battery?
 
You must log in or register to reply here.
Back
Top